Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

One recurring theme in my research and writing on agentic AI security has been the distinction between soft guardrails and hard boundaries. As someone who serves on the Distinguished Review Board for the OWASP Agentic Top 10, and who spends every day thinking about how to secure agents across enterprise environments at Zenity, this distinction is not academic. It is potentially the single most important conceptual framework practitioners need to internalize right now.

AI agents are rapidly becoming embedded in enterprise workflows, influencing revenue operations, customer engagement, development, and internal decision-making. As these systems gain autonomy and inherit access across SaaS, cloud, and endpoint environments, they introduce a new layer of operational and security risk that traditional controls cannot fully manage.

How a routine calendar invite enabled silent local file access and data exfiltration Note: This post is part of a coordinated disclosure by Zenity Labs detailing the PleaseFix vulnerability family affecting the Perplexity Comet Agentic Browser. This blog focuses on browser-level autonomous agent execution and session compromise.

If you’re like me, you’ve been enthralled with the recent story, expertly written by Sean Hollister at The Verge, about how Sammy Azdoufal built a remote control for his DJI Romo vacuum with a PlayStation controller, and ended up in control of 7,000+ robovacs all over the world. On the surface, it sounds like vibe coding gone slightly sideways. I mean, really, what could a vacuum possibly do? Turns out… a lot.

In my previous piece, "The Agentic AI Governance Blind Spot," I laid out what I believe is one of the most critical gaps in the AI governance landscape today: the three most cited frameworks in AI governance, NIST AI RMF, ISO 42001, and the EU AI Act, don’t contain a single mention of agentic AI. Not one reference to autonomous agents, multi-agent systems, or AI that takes actions with real-world consequences. The response to that piece confirmed what I suspected.

OpenClaw exposes a fundamental misalignment between how traditional enterprise security is designed and how AI agents actually operate. As an AI agent assistant, OpenClaw operates with human permissions, executes actions autonomously, and processes untrusted content as input, all while sitting outside the visibility of conventional security tools.

Approach any security, technology and business leader and they will stress the importance of governance to you. It’s a concept echoed across board conversations, among business and technology executives and of course within our own echo chamber of cybersecurity as well. For example, the U.S. Cybersecurity Information Security Agency (CISA) has a page dedicated to Cybersecurity Governance, which they define as.

Today we’re excited to announce that Zenity now protects the most powerful, enterprise-critical coding assistants - Cursor, Claude Code, and GitHub Copilot - from build-time to runtime. As AI becomes a first-class developer tool, Zenity gives security teams the visibility and control they need to safely embrace coding assistants everywhere they’re used, in IDEs, CLIs or in the cloud.

Security teams are entering a new phase of risk driven by the combination of AI agents and broad access to internal and external data. Agents are no longer limited to responding to prompts. They read files, pull documents from shared repositories, query external sources, and move information across systems on behalf of users. This shift brings real business value. Knowledge becomes easier to access, workflows move faster, and information that once required deliberate effort can be surfaced instantly.

In the first round of the AI gold rush, most conversations about AI security centered on models: large language models, training data, hallucinations, and prompt safety. That focus made sense when AI was largely confined to generating text, images, or recommendations. But that era is already giving way to something far more consequential.